To achieve accurate prediction and effective control of retaining wall deflection in soft soil excavations and to ensure construction safety, this study developed a spatiotemporal distribution matrix of retaining wall displacement based on its significant spatiotemporal distribution characteristics and created a hybrid CNN-LSTM prediction model that integrates convolutional neural networks (CNN) and long short term memory (LSTM). The research synchronously forecasted and compared the wall deflection from both temporal and spatial dimensions through a deep excavation project in Shanghai. The results show that: 1) Compared with four traditional conventional models, the CNN-LSTM hybrid prediction model, based on the spatiotemporal distribution matrix of the retaining wall displacement, demonstrates precise prediction of the spatiotemporal distribution of horizontal displacement through the extraction of spatiotemporal distribution characteristics and deep learning. 2) For spatial distribution prediction, the extraction of spatial distribution features combined with deep learning enables not only accurate identification of the deflection mode of the retaining wall but also precise prediction of distribution features such as deformation curvature and positions corresponding to maximum deflection. The predicted MAE in depth and horizontal directions is 0.532 mm and 0.742 mm, respectively. 3) For time distribution prediction, the dynamic forecasting of retaining wall displacement at various construction stages is achieved through the extraction of horizontal displacement time series features and deep learning, while considering both short- and long-term data dependencies. The predicted results during the construction period demonstrate good robustness, with an MAE of 0.841 mm.
为验证CNN-LSTM模型在挡墙变形时空分布预测方面的效果及优势,将CNN-LSTM与4种常用预测模型进行比较,包括CNN、LSTM、ANN和卷积长短时记忆网络(convolutional-long short term memory,Conv-LSTM)模型. 5种模型的训练与预测误差如 表6所示.
LIAOS M, FANY Y, SHIZ H,et al .Optimization study on the reconstruction and expansion of an underground rail transit center in Shanghai soft ground[J].Tunnelling and Underground Space Technology,2013,38:435-446.
[2]
TANY, LIM W .Measured performance of a 26 m deep top-down excavation in downtown Shanghai[J]. Canadian Geotechnical Journal,2011,48(5):704-719.
LIH, LIZ W, LIAOS M,et al .Field measurement of time-space distribution behaviors of environmental settlement of an ultra-deep excavation in Shanghai soft ground[J]. Chinese Journal of Geotechnical Engineering,2023,45(8):1595-1604.(in Chinese)
XUZ H, WANGJ H, WANGW D .Deformation behavior of diaphragm walls in deep excavations in Shanghai[J].China Civil Engineering Journal,2008,41(8):81-86.(in Chinese)
LIAOS M, WEIS F, TANY, et al .Field performance of large-scale deep excavations in Suzhou[J]. Chinese Journal of Geotechnical Engineering, 2015, 37(3): 458-469.(in Chinese)
[9]
YANGT, LIUS L, WANGX Y,et al. Analysis of the deformation law of deep and large foundation pits in soft soil areas[J].Frontiers in Earth Science,2022,10:828354.
[10]
LEIG, GONGX N .Analysis of lateral displacement law of deep foundation pit support in soft soil based on improved MSD method[J].Advances in Civil Engineering,2021,2021:5550214.
[11]
DONGY, LUANY Z, WANGF,et al .Monitoring and prediction of horizontal displacement of underground enclosure piles in subway foundation pits[J].ACS Omega,2023,8(26):23389-23400.
FENGJ F, YUJ L, YANGX L,et al .Back analysis and prediction of deep pit foundation excavation considering dynamic factors[J].Rock and Soil Mechanics, 2005, 26(3): 455-460.(in Chinese)
YANGL D, ZHONGC G, ZENGJ L. Dynamic prediction for displacement and safety of foundation pit[J]. China Civil Engineering Journal,1999,32(2):9-13.(in Chinese)
YANGL D, QIUS H, YANGZ X .Prediction on displacement and stability of frame bracing structure of foundation pit[J].Rock and Soil Mechanics,2001,22(3):267-270.(in Chinese)
[18]
DANK, SAHUR B. Estimation of ground movement and wall deflection in braced excavation by minimum potential energy approach[J]. International Journal of Geomechanics, 2018, 18(7): 04018068.
[19]
TONGL H, GUOW L, XUC J,et al .Simplified theoretical prediction for lateral deformation of a diaphragm wall using the general third-order plate theory[J]. International Journal of Geomechanics,2023,23(11):04023188.
[20]
DOT N, OUC Y, CHENR P .A study of failure mechanisms of deep excavations in soft clay using the finite element method[J].Computers and Geotechnics,2016,73:153-163.
[21]
HASHASHY M A, WHITTLEA J .Ground movement prediction for deep excavations in soft clay[J].Journal of Geotechnical Engineering,1996,122(6): 474-486.
[22]
ZHANGW G, GOHA T C, XUANF .A simple prediction model for wall deflection caused by braced excavation in clays[J].Computers and Geotechnics,2015,63:67-72.
TANY H, JIT J .Analysis and calculation of lateral wall deflection of braced excavation in soft-clay[J].Chinese Journal of Geotechnical Engineering,1995,17(4):71-76.(in Chinese)
JINX F, LIANGS T, ZHUX J,et al .Simplified method for calculating maximum deformation of diaphragm walls caused by braced excavation in soft clays[J].Rock and Soil Mechanics,2015,36(Sup.2):583-587.(in Chinese)
LIUG B, SHENJ M, HOUX Y .Reliability analysis of braced structures of deep-excavation[J].Journal of Tongji University (Natural Science),1998,26(3):260-264.(in Chinese)
LIUY, FENGZ, HUANGG C,et al .The study in predicting the deformation of supporting structure for deep foundation pit[J].Chinese Journal of Underground Space and Engineering, 2009, 5(2):329-335.(in Chinese)
ZHANGY L, NIEZ Y, LIF X, et al. Deformation prediction of excavations based on numerical analysis[J]. Chinese Journal of Geotechnical Engineering,2012,34(Sup.1):113-119.(in Chinese)
[33]
WANGD Y, WUK P, WANGJ,et al .Deformation monitoring and simulation analysis of deep foundation excavation construction for subway station[J].IOP Conference Series:Earth and Environmental Science,2020,580(1):012027.
[34]
BHATKART, BARMAND, MANDALA,et al .Prediction of behaviour of a deep excavation in soft soil:a case study[J].International Journal of Geotechnical Engineering,2017,11(1):10-19.
[35]
XIAOH H, CAOZ Y, CAOR L,et al .Prediction of shield machine posture using the GRU algorithm with adaptive boosting:a case study of Chengdu Subway project[J].Transportation Geotechnics,2022,37:100837.
[36]
BAID X, LUG Y, ZHUZ Q,et al .Using time series analysis and dual-stage attention-based recurrent neural network to predict landslide displacement[J].Environmental Earth Sciences,2022,81(21): 509.
[37]
YANGY F, LIAOS M, LIUM B .Dynamic prediction of moving trajectory in pipe jacking:GRU-based deep learning framework[J].Frontiers of Structural and Civil Engineering,2023,17(7):994-1010.
[38]
ZHANGP, WUH N, CHENR P,et al .A critical evaluation of machine learning and deep learning in shield-ground interaction prediction[J].Tunnelling and Underground Space Technology,2020,106:103593.
QIANJ G, WUA H, JIJ,et al .Prediction for nonlinear time series of geotechnical engineering based on wavelet-optimized LSTM-ARMA model[J].Journal of Tongji University (Natural Science),2021,49(8):1107-1115.(in Chinese)
[41]
WUC Z, HONGL, WANGL,et al .Prediction of wall deflection induced by braced excavation in spatially variable soils via convolutional neural network[J].Gondwana Research,2023,123:184-197.
[42]
YONGW X, ZHANGW G, NGUYENH,et al .Analysis and prediction of diaphragm wall deflection induced by deep braced excavations using finite element method and artificial neural network optimized by metaheuristic algorithms[J]. Reliability Engineering & System Safety,2022,221:108335.
[43]
TRAND, NGUYENH, WANGY R,et al .Analysis of artificial intelligence approaches to predict the wall deflection induced by deep excavation[J].Open Geosciences, 2023, 15: 20220503.
XUC J, LIX Y .Lateral deformation prediction of deep foundation retaining structures based on artificial neural network[J].Journal of Shanghai Jiao Tong University,2024,58(11):1735-1744.(in Chinese)
[46]
ZHAOH J, LIUW, SHIP X,et al .Spatiotemporal deep learning approach on estimation of diaphragm wall deformation induced by excavation[J].Acta Geotechnica,2021,16(11): 3631-3645.
[47]
KUNGG T C, HSIAOE C L, SCHUSTERM,et al .A neural network approach to estimating deflection of diaphragm walls caused by excavation in clays[J].Computers and Geotechnics,2007,34(5):385-396.
[48]
YANGY F, LIAOS M, TANGL H .Physics-guided architecture of neural networks for predicting wall deflection induced by braced excavations[C]//Information Technology in Geo-Engineering.Cham:Springer,2025:295-304.
[49]
YANGY F, LIAOS M, TEOHB K,et al .A physics-constrained neural network for predicting excavation-induced ground surface settlement in clay[J]. Journal of Rock Mechanics and Geotechnical Engineering,2025,17(5):2665-2681.
[50]
LIUY B, LIAOS M, YANGY W,et al .Data-driven and physics-informed neural network for predicting tunnelling-induced ground deformation with sparse data of field measurement[J]. Tunnelling and Underground Space Technology,2024,152:105951.
LIUJ H, LIUG B, FANY Q. The theory and its practice by using the rule of time-space effect in soft soil excavation(part1)[J]. Undergrourd Engineering and Tunnels,1999(3):7-12.(in Chinese)
LIJ C, ZHANGZ Y, LUOG Y .Study on effects of time-space of retaining structures of the deep-foundation pit excavation[J].Rock and Soil Mechanics,2003,24(5):812-816.(in Chinese)
[55]
FINNOR J, BLACKBURNJ T, ROBOSKIJ F. Three-dimensional effects for supported excavations in clay[J]. Journal of Geotechnical and Geoenvironmental Engineering,2007,133(1):30-36.
LIL, YANGM, XIONGJ H. Analysis of the deformation characteristics of deep excavations in soft clay[J]. China Civil Engineering Journal,2007,40(4):66-72.(in Chinese)
[58]
DONAHUEJ, HENDRICKSL A, GUADARRAMAS,et al .Long-term recurrent convolutional networks for visual recognition and description[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 7-12,2015,Boston,MA,USA.IEEE,2015:2625-2634.
[59]
MAHALANOBISP C .On the generalized distance in statistics[J]. Sankhyā: The Indian Journal of Statistics, 2018, 80: S1-S7.
[60]
LIJ H, LIP X, GUOD,et al .Advanced prediction of tunnel boring machine performance based on big data[J]. Geoscience Frontiers,2021,12(1):331-338.
[61]
LEE RODGERSJ, NICEWANDERW A .Thirteen ways to look at the correlation coefficient[J]. The American Statistician,1988,42(1): 59-66.